Download PARTITIVE RESTRICTIVE mODIFICATION OF NAmES IN

Quaderns de Filologia. Estudis lingüístics. Vol. XIV (2009) 33-56
Partitive restrictive modification
of names in English: Arguments for
their metonymic motivation
Antonio Barcelona Sánchez
Universidad de Córdoba
0. Introduction1
The standard morphosyntactic constraints in English on those names that
consist of a single proper noun prevent the combination of plural morphemes,
determiners and restrictive modification with them:
(1) Paris - *the Paris - *a Paris - ?I like the beautiful Paris, but not the ugly
one.
The same limitations apply to the combinations between proper nouns and
other words that result in complex names. Take The Hague:
(2) The Hague - *Hague - *a Hague - * I love the good Hague, and I hate the
bad Hague
These limitations are due to the fact that names typically have unique
reference. However, standard grammars, like Quirk et al. (1985: 288-297) or
Jespersen (1949: 426-429, 488-491, 544-580), in fact devote plenty of space
to the numerous cases in which the above constraints are violated. In several
papers (Barcelona, 2003b; 2003c; 2004), I have dealt briefly with several of
these cases, particularly with the type of constraint violation consisting in
the use of proper nouns as “paragons” (Lakoff, 1987), as in That student is
a Shakespeare. In this paper, I will concentrate on another type of violation
of these constraints, namely “partitive restrictive modification” on proper
nouns, and I will argue for its metonymic motivation. But, before discussing
this claim, I find it necessary to characterize briefly the notion of metonymy
The research reported in this article is part of a research project funded by the Spanish Government,
Ministerio de Ciencia e Innovación (Ministry of Science and Innovation), Project no. FFI200804585.
1
34
Antonio Barcelona Sánchez
assumed in this paper (section 1), and to show that the metonymic motivation
of grammatical phenomena, rather than being exceptional, is, in fact, quite
frequent (section 2).
1. Notion of metonymy
The concept of metonymy assumed in the present paper is similar to
Kövecses and Radden’s (Kövecses and Radden, 1998: 39), here reproduced
as (3):
(3) Metonymy is a cognitive process in which one conceptual entity,
the vehicle, provides mental access to another conceptual entity,
the target, within the same domain, or ICM.
In Barcelona (2003a) I proposed to distinguish, for various research purposes,
four different general classes of metonymies which progressively constrain
their range of membership. Schematic metonymies are those that satisfy the
minimal requirements for every conceptual metonymy: intra-domain mapping
and activation of target by source, irrespective of the neat distinctiveness of
source vis-à-vis target. Typical metonymies are those schematic metonymies
in which the source and the target are neatly distinct (within the super-ordinate
overall domain) from each other. Prototypical metonymies are those typical
metonymies with individuals as targets and as referents (they are the “classical”
instances of metonymy). Conventional metonymies are typical metonymies
which are socially sanctioned in virtue of a number of parameters (see below).
In order to explore the grammatical relevance of metonymy it is convenient
to assume as broad a notion of metonymy as possible, so that we can detect
its presence in the many grammatical structures that can be motivated by it.
Therefore, by “metonymy” I will be assuming in the present article the first of
these classes. This is the definition offered in Barcelona (2002: 246).
(3) A schematic metonymy is an asymmetric mapping of a conceptual domain,
the source, onto another domain, the target. Source and target are in the same
functional domain and are linked by a pragmatic function, so that the target is
mentally activated.
As I stated earlier, this definition and Kövecses and Radden’s have
several points in common. Both are fairly broad definitions of metonymy,
which highlight the fact that metonymy is fundamentally a “reference-point”
phenomenon (Langacker, 1993: 29-35). But there are also some significant
differences between them. I prefer to use the term source (rather than vehicle).
Partitive restrictive modification of names in English...
35
For reasons discussed in Barcelona (2003a) I define metonymy as a conceptual
projection or mapping, as claimed by Lakoff and Turner (1989: 103); however,
whereas metaphor consists of a set of structural, symmetric sub-mappings
or correspondences (a systematic matching of counterparts between source
and target), metonymy is a single, asymmetric mapping, with no counterpart
matching. I prefer the term domain instead of conceptual entity because even
“entities” are understood in terms of complex domain matrices (Langacker,
1987: chapter 4), and because when metonymy is involved in category
structure, it is complex categories, rather than individual entities, that are
sources and targets–just think of the stereotype-based metonymic model of
“bachelorhood” or “motherhood” (Lakoff, 1987: chapter 8). However, there
is no harm in talking about metonymic sources or targets being “entities” in
cases in which they are clearly unitary physical objects (including people) with
a given role, or their clearly delineated parts2. In Barcelona (2002) I specify
that the overall domain where metonymy occurs is a “functional domain”,
i.e. a “frame” (Fillmore, 1985) or an “I.C.M.” (Lakoff, 1987), and that source
and target must be linked by a “pragmatic function” (Fauconnier, 1997), i.e.
the role of the source is closely connected in experience to that of the target
(author-work, image-object, etc.). The other differences are simply a matter
of personal preference.
On the current debate in cognitive linguistics regarding the nature of
metoymy, see Benczes, Réka; Barcelona, Antonio and Ruiz de Mendoza
Ibáñez, Francisco-José, eds. (in preparation), with contributions from most of
the leading authors in the field.
2. Metonymy in grammar
Metonymy is regarded by cognitive linguists as a fundamental cognitive
model, together with metaphorical, image-schematic and propositional
models (Lakoff, 1987: 77-90, 154). All of these models play a crucial role in
the grammar of languages, as cognitive linguistics has shown. The study of
the role of metonymy in grammar relevance is one of the most recent areas
of has only recently begun to be explored, but there is already a substantial
body of research demonstrating the metonymic basis of a large number of
grammatical phenomena. Langacker (1999: 67) states that though “usually
regarded as a semantic phenomenon, metonymy turns out to be central and
essential to grammar”, and that grammar is “a rich source for the investigation
of metonymy. At the same time, a recognition of its prevalence and centrality
For instance, president (a person
person, as in We need strong arms.
2
with a role) for army, as in Nixon bombed Hanoi, or arm for
36
Antonio Barcelona Sánchez
is critical not just for describing grammar but for a realistic assessment of its
basic nature”. Langacker (in preparation), furthermore, adds that grammar is
“basically metonymic, in the sense that the information explicitly provided
by conventional means does not itself establish the precise connections
apprehended by the speaker and hearer in using an expression”; that is,
grammar only provides a blueprint of that information.
Among the many studies on the interaction of grammar and metonymy, we
can only mention the following (the list is simply illustrative):
– The metonymic nature of a great many instances of active zone / profile
discrepancy (Langacker, 1993: 33-35; 1999: 200-201; in preparation):
topic constructions, certain relational subjects or objects (like a novel,
standing for the relation “X write / read a novel” in Zelda began a
novel), certain descriptive adjectives (like fast in Therese is fast),
“raising” in general (Langacker, 1991: 189-201; 1995; 1999: 317-360),
and relative-clause constructions. Brdar and Brdar-Szabó (2004) is an
excellent paper on the cross-linguistic differences between English
constructions with metonymy-based “raising” adjectives and the
corresponding constructions in other languages.
– Anaphora resolution (Langacker, 1999: Ch. 7 and 9; in preparation;
Ruiz de Mendoza & Otal, 2002).
– Various aspects of clausal constructions (Caballero, in preparation;
Ziegeler and Lee, in preparation; Brdar-Szabó, in preparation; Peña and
Ruiz de Mendoza, in preparation).
– Certain types of conversion (Dirven, 1999; Kövecses and Radden,
1998: 60 commenting on Clark and Clark, 1979; Radden, 2005: 16-17;
Barcelona, 2004; in preparation a, in preparation c)
– Other word-formation processes, including certain types of derivation
and compounding (Panther and Thornburg, 2002; Radden, 2005: 1720; Barcelona, 2008 on bahuvrihi compounds; Basilio, in preparation;
Palmer, Rader and Clarito, in preparation).
– Aspectual shift (Panther and Thornburg, in preparation) and the use of
stative lexemes (verbs, adjectives, etc) in dynamic contexts (Panther
and Thornburg, 2000).
– Proper names and noun phrases (Brdar & Brdar-Szabó, in preparation;
Brdar (2007a), including generic NPs (Radden, in preparation).
– Grammaticalization processes (e.g. Traugott and Dasher, 2002;
Guarddon, in preparation).
Partitive restrictive modification of names in English...
37
Two useful surveys on the wide-ranging interaction between metonymy
and grammar are Ruiz de Mendoza and Otal Campo (2002) and Brdar (2007b).
See also Panther, Thornburg & Barcelona, eds. (in preparation) and Barcelona
(in preparation a, in preparation c).
3. The “irregular” behavior of names in two standard grammars
This issue is dealt with by three standard grammars: Greenbaum and Quirk
(1990), Quirk et al. (1985), and Jespersen (1949). Many of the observations
presented in the older handbook are incorporated by the two later books, which
deal systematically with this issue. Their discussion of the various types of
violations of the grammatical constraints on names is summarized as below
(the examples are drawn indistinctly from either book)3:
(a) Different referents with the same name: There is a Richmond in the south
of England and a Richmond in the north, not to mention a dozen Richmonds
outside the British Isles (places with the same name) / / The Dr Brown I know
comes from Ireland, not from Scotland.
No generic statements are possible with these cases: *Richmonds are
always splendid for a vacation.
(b) Different referents with the same name, which also constitute a
conceptual class. For example, names of week days constitute a class in virtue of
the properties they share (Mondays are the first working day of the week; every
Christmas shares with the others even more sharply defined characteristics):
She always spends her Mondays / Octobers / Christmases here. The same
applies to names of months and festivals,
(c) Uses due to “informal conventions”:
(c-1) Married couples and families: The Wilsons.
(c-2) Use of a famous name to mean the type that made it famous:
There were no Shakespeares in the ninenteenth century [example paraphrased
by Quirk & Greenbaum as “...writers who towered over contemporaries
as William Shakespeare did over his”, and by Quirk et al as “authors like
Shakespeare”].
(d) Names subject to modification:
(d-1) Uses due to what these grammars call “partitive restrictive
modification”: In the England of Queen Elizabeth / The Chicago I like / The
The 1990 grammar attempts to explain some of the irregularities in conceptual terms. Therefore,
I reflect below the general classification provided by this grammar, adding to it some of the more
relevant remarks in the 1985 handbook.
3
38
Antonio Barcelona Sánchez
young Joyce already showed signs of the genius that was to be fulfilled in
Ulysses.
(d-2) Uses due to nonrestrictive modification, either colloquial and
stereotyped (Poor old Mrs Fletcher) or formal and often stylized (The fondly
remembered John F. Kennedy).
(e) Miscellaneous: A Rolls Royce/This museum has several Renoirs/A pack
of Chesterfields.
The term “partitive restrictive modification” is not explained by Greenbaum
and Quirk (1990). Quirk et al (1985: 290) explain it by saying that “cataphoric
the with restrictive modification can have the effect of splitting up the unique
referent of the proper noun into different parts or aspects” [my italics].
Therefore, they classify this as a type of partitive meaning.
Both grammars say that in all of the cases in table 1 names are transiently
reclassified as common nouns4. Jespersen (1949: 426, 544, 573-577) records
the uses in paragraph (c-2) of the table, but not instances of partitive restrictive
modification.
Apart from the instances listed above, the general “rules” for the use or
omission of the definite article with certain names (e.g. with geographical and
other names, etc.) are discussed extensively by the three standard grammars.
We have no space to deal with them here, although a conceptual motivation
could probably be discerned for many of them, too (see, in this respect, Radden
and Dirven, 2007: Ch. 5).
An initial published attempt at a conceptual explanation of all of these
“irregular” uses in terms of conceptual metonymy is Barcelona (2003b) and in
Barcelona (n.d.) The uses referred to above as c-2 are also known as “paragon”
uses of names (Lakoff, 1987: 87-88). I offered a metonymic account both of
these uses and of partitive restrictive modification in a crosslinguistic study in
Barcelona (2003c). And in Barcelona (2004) (written at about the same time
but published later) I carried out a detailed investigation of the metonymic
motivation for the use of names as paragons in English.
As stated in section 1, the present paper is an in-depth discussion of the
metonymic motivation of partitive restrictive modification of names in English
(the uses listed above as d-1). The discussion in the following section greatly
This transient reclassificaction occasionally becomes permanent. These are just some examples:
a mackintosh ‘a type of raincoat’ (derived from Mackintosh, the inventor of this type of raincoat:
inventor for invention, a subtype of author for work); a Xerox ‘a photocopy’ (from Rank Xerox,
the company manufacturing the first photocopying machines: producer for product and instrument
for object involved in action); a Jeremiah ‘someone characterised by making gloomy predictions’
(from Jeremiah, the prophet, a paragon of this type of people: characteristic property for entity and
stereotypical member for category; see Barcelona, 2004).
4
Partitive restrictive modification of names in English...
39
expands on and specifies the claims made in Barcelona (2003b) and Barcelona
(2003c).
4. Metonymic
analysis of non-paragon names with partitive restrictive
modification
I offer below a somewhat clearer classification of the instances of names
subject to modification enumerated in paragraph (d) of the previous section.
Names subject to modification can be of two main kinds:
(a) Names subject to non-restrictive modification: Poor old Mrs. Fletcher
is standing in her kitchen.
(b) Names subject to restrictive modification, with two further subtypes:
(b-1) Non-partitive: The Dr Brown I know comes from Ireland, not
from Scotland. This type of restrictive modification is often associated with
special use (a) in section 3 (i.e. different referents with the same name).
(b2) Partitive: The young Joyce was already a great writer (referring
to the writer as a young man).
In Barcelona (2003b), and Barcelona (n.d.) strong arguments were presented
for the claim that both the non-restrictive modification (a) and the non-partitive
restrictive modification (b-1) of names is motivated by metonymy. In the rest
of this section, therefore, I concentrate on the main topic of the paper, namely
the conceptual motivation of partitive restrictive modification.
Partitive restrictive modification, like all of the “irregular” uses of names
listed in section 3 (especially their use as paragons; Barcelona (2003b,
2004; Brdar 2007a) is evidence that names are often understood against the
background of an experience-based conceptual network, which is brought
to bear on their interpretation and their use. In this respect, then, they are
not really different from common nouns. The main difference is that names
normally have unique reference, whereas common nouns must enter phrasal
referring expressions to be referential. However, the conceptual network
underlying a name can be invoked, and certain “aspects or circumstances”, that
is, “sub-domains”, included in it can be focused on and singled out. In the use
under discussion, the referent of the name, which is regularly construed as a
unitary entity, is figuratively “split up” and re-categorized as a class of entities.
How is this partition achieved? In my view, at least the roots of the partition
are metonymic. The following is one of the examples of partitive restrictive
modification borrowed from Greenbaum & Quirk (1990: 88):
40
Antonio Barcelona Sánchez
(4) The young Joyce already showed signs of the genius that was to
be fulfilled in Ulysses.
In this example, the prominent (sub)domain in the knowledge network
associated with the individual James Joyce, the writer, is that of age scale
(youth in this case). More precisely, the domain associated with joyce that
ultimately becomes “active” as a mental space (Fauconnier, 1997) in this
example, is the whole domain of Joyce’s biography, activated by age scale,
the latter being activated in turn by the direct activation of one of the subdomains of this age scale: youth. In other words, the example could be roughly
paraphrased as As a young man (youth ⇒ age scale ⇒ biography), Joyce
already showed signs... This “activation chain” can be claimed to be itself
metonymic, with each domain successively activating (as parts) the domain
including them (youth presupposes an age scale and jointly they presuppose
biography, etc.). But the activation chain facilitates the really relevant factor
motivating the use of this partitive restrictive modification, namely the active
zone metonymy which leads to the mapping of the cognitive domain joyce
onto one of its possible active zones: Joyce when young. This active zone
constitutes a relation involving joyce (i.e. the relation is the predication of
youth on Joyce) which is situated at a time period determined by the active
mental space Joyce’s biography. This time period does not coincide with
speech or writing time (i.e the time of uttering example (4), if it was part of a
spoken utterance, or with the time of writing the corresponding sentence, if it
was part of a written text): It is necessarily prior to it, due not only to the past
tense of the verb showed but also, and fundamentally, to the currently active
mental space, which necessarily dicusses past situations.
The noun phrase The young Joyce in this sentence is, thus, understood as
referring to “Joyce at the time when he was young”, not to a neutral, holistic
notion of Joyce independent from any specific circumstances affecting him.
Note that the paraphrase provided by Greenbaum and Quirk for this example
is ‘Even while he was young, James Joyce...’.
4.1. A summary of the properties of active zone metonymies
In order to understand this claim properly, we should first characterize in
some detail the notion of “active zone metonymy”, as put forward by Langacker.
An “active zone” can be a portion of the entity profiled (in Langacker’s sense
of “profile”) by a linguistic expression; this portion is crucially and directly
involved in a relationship in which the entity as a whole participates. In
Langacker’s well-known example Your dog bit my cat, “certain portions of
Partitive restrictive modification of names in English...
41
the dog5 (notably the teeth and jaws) are directly and crucially involved in the
biting, and others (e.g. the tail and pancreas) hardly at all” (Langacker, 1999:
62). And only some unspecified portion of the cat suffers the biting.
But an active zone does not have to be a sub-part of the entity conventionally
profiled by a linguistic expression, as in the above metonymic uses of dog and cat.
It may be just another entity or relationship closely associated in encyclopedic
knowledge with the profiled entity. Langacker says that the active zone may
simply be located in the profiled entity’s “dominion” (Langacker, 1999: 200201), which is equivalent to a “domain” - in the sense of “domain” used in
this paper and in the standard cognitive linguistic definitions of metaphor and
metonymy (e.g. Kövecses and Radden, 1998, Lakoff and Turner, 1989: 103).
In more precise terms, we can say that the active zone may simply be located
in the frame associated to the profiled entity.
Active zone/profile discrepancy is, according to Langacker, a type of
metonymy, ubiquitous and essential to language (1999: 67, 200, 331)6.
Langacker makes the important point that active zone / profile discrepancy is
the normal situation in cognition and in language (1999: 62-63), there being
very few cases in which such discrepancy does not operate. He also claims
that resorting to periphrases mentioning the active zone is “otiose unless they
provide information not otherwise available” (1999: 65). An example is She
blinked her big blue eyes, where explicit mention of the eyes as the object
nominal referring to the subject’s active zone with respect to blinking, “allows
the specification of their properties” (1999: 64).
A fundamental consequence of the flexibility in active zone selection is
that an active zone can be another relationship in which a profiled participant
is involved. In these cases, the occurrence of active zone / profile discrepancy
can actually be predicted, because “a relation is intrinsically associated with
its participants, which usually have greater cognitive salience (owing to
concreteness or to their nominal character)” (1999: 66), and so constitute
natural metonymic “reference points” for the relationships in which the
participants are involved7. Some examples were cited earlier: sentences like
Zelda began a novel, Therese is fast. In raising constructions, the additional
This type of animal is the entity holistically profiled in the conventional prototypical meaning of the
lexeme dog. And a token of this type is the entity holistically profiled in the prototypical meaning of
the noun phrase your dog.
5
On the metonymic status of active zone/ profile discrepancy, see Paradis (2004; in preparation), who
denies it any metomymic status; and Peirsman and Geeraerts (in preparation) and Barcelona, who
claim that certain instances of zone activation are clearly metonymic while others are doubtfully so.
6
Langacker claimed long ago (see 1993: 29ff) that metonymy in general is a type of reference point
phenomenon, with the source functioning as a reference point.
7
42
Antonio Barcelona Sánchez
relationship constituting the active zone of the participant coded as subject
or object is often explicitly mentioned because it provides information not
otherwise available. For instance, in Mary is easy to please, the active zone of
the subject nominal Mary is the relationship coded by the infinitive clause. The
explicit mention of the active zone is necessary, because the speaker intends to
specify that the participant Mary can be placed on a scale of relative difficulty
only with respect to the relationship “X please Mary”. Other examples are
Fred is really slow at paying his debts or When it comes to paying his debts,
Fred is really slow: “it is only by engaging in some process, in this case pay his
debts, that an individual can be located on a scale of relative speed” (Langacker
1999: 66).
4.2. The active zone analysis of partitive restrictive modification
Now we are in a position to begin to understand why partitive restrictive
modification can be claimed to be conceptually motivated by an active zone
metonymy.
This type of modification on names is motivated by a metonymic mapping
of a given participant in the overall relationship coded by the clause onto an
active zone of that participant. In example (4), the speaker maps the participant,
in this case the unique individual joyce, onto one of the possible active zones
of the latter, namely a temporally earlier relationship in which this participant
was involved: that of Joyce when young. This relationship is one of the
many circumstances affecting this participant (in this case, one of the many
biographical circumstances affecting him), all of them part of his “dominion”,
that is, part of the domain joyce. In other words, we have a discrepancy between
the profile of the name Joyce (which profiles an individual “in toto”) and its
active zone. See Figure 1
Figure 1: Metonymic motivation of the partitive restrictive
modification in example 4.
JOYCE FRAME
JOYCE’S BIOGRAPHY
AGE
JOYCE WHEN YOUNG
Partitive restrictive modification of names in English...
43
The active zone is made partially explicit in this example thanks to the
modifier young. The reason for doing so is that this mention provides information
not otherwise available. Note that the speaker might not have made the active
zone explicit, among other reasons because he (perhaps wrongly) had assumed
that his listeners could figure out (on the basis of context and / or general
background knowledge) the relevant active zone of joyce in this example. For
instance, he might just have said Joyce already showed signs of the genius
that was to be fulfilled in Ulysses, meaning, in fact, something like When he
was young, Joyce already..., and leaving the active zone of Joyce implicit.
The active zone of the subject with respect to “being fast” would also remain
implicit in a sentence like Therese is fast, but it might also be recovered from
context (e.g. “fast at typing”)8. But this does not mean that the choice of active
zone is unconstrained in this case. The complex relation “Joyce show signs of
the genius that was to be fulfilled in Ulysses” is temporally grounded in the
past9, specifically in a period in Joyce’s biography taking place before an event
(Joyce’s genius being fulfilled in Ulysses) which is known to have occurred
in a later period of his biography. This specific grounding of the complex
relationship profiled by the clause constrains the choice of other relationships
as active zone interpretations of its main participant (“Joyce”): They must have
the same temporal grounding. That is, given an adequate knowledge of Joyce’s
biography, in the sentence Joyce already showed signs of the genius that was
to be fulfilled in Ulysses, listeners can understand by Joyce “Joyce as he wrote
Dubliners”, “Joyce in 1910” (he was born in 1882), “Joyce when young”,
etc., but not “Joyce when middle-aged”, or “Joyce as he wrote Finnegan’s
Wake”. Since this constraint does not reduce the range of options to just one,
the speaker may still have a communicative motive to unequivocally single out
one active zone. In example (4), the motive is to contrast not just Joyce’s early
work with his later work, but also to highlight the fact that he was a young man
when he produced this early work.
It might be objected that the name Joyce in (4) simply stands for that
individual, not for that individual as affected by any circumstance, and that it is
An explanation for the need to mention explicitly in (4) the active zone of the profiled entity Joyce
with respect to the relationship “Joyce showing signs of the genius... etc” may be that, contrary to
the normal situation (see Langacker, 1999: 62-67, etc.), such circumstances as age or profession, are
not automatically understood as active zones of the profiled entity with respect to a relationship in
which it may be involved. Compare with the automatic mapping (with minimal contextual support)
of participants onto their body parts as active zones (a. z.) with respect to such relationships as biting
(a. z.: mouth), hitting (a. z.: fist, hand), kicking (a. z.: foot), etc.: The dog bit me, The crook hit him
hard, Mike kicked the door shut.
8
The term “ground” in Cognitive Grammar refers to the speech event and its participants, which can
be indicated explicitly by means of deictic elements, among them, tense morphemes, determiners,
etc., or remain implicit.
9
44
Antonio Barcelona Sánchez
simply the addition of young that activates a temporally bounded circumstance,
as a sort of space-builder (Fauconnier, 1997: 40). Thus no metonymic analysis
of Joyce would be necessary. My reply to this objection is that, even if I
absolutely agree that young is a space-builder, Joyce is still metonymic in this
example. That this is so can be seen by pointing out the difference between
example (4) and example (5):
(5) The young Joyce is more intelligent than the old Joyce.
The noun Joyce in example (4) has exclusively an active zone reading if
the speaker is talking in the year 2009 about James Joyce, the famous Irish
novelist, and if both speaker and listener(s) share a basic knowledge of Joyce’s
biography. That is, the speaker can only be talking about James Joyce the
writer when he was a young man, implicitly contrasting, as we said above, his
literary skills with those he displayed at a later stage in his life. On the other
hand, each of the two occurrences of the noun Joyce in example (5) would
be interpreted in the normal profile of the noun, i.e. as holistically profiling a
different unique individual in each case: The speaker might be talking in the
year 2009 about two different members of the Joyce family, and contrasting
them in terms of their intelligence. In example (5) we have an instance of
the special use due to different referents having the same name (use a; see
section 3); in this case the names have non-partitive restrictive modification to
distinguish one referent from the other (see the beginning of section 4 on “nonpartitive restrictive modification”). The conceptual motivation of this type of
examples (this motivation being a different kind of metonymy, namely name
for individual) was discussed in Barcelona 2003b.
As we can see, in both examples, as in most linguistic expressions, context
and background knowledge are essential to arrive at the intended interpretation;
in this case, to carry out either an active zone or a holistic construal of the noun
Joyce.
The use of the modifier and the determiner in example (4) is triggered
and made possible by the speaker’s previous prelinguistic mental metonymic
mapping of “the whole” of joyce frame onto an active zone consisting of a
specific past relation in which Joyce was involved (“Joyce when young”). As
Langacker says (1999: 67), “a variety of grammatical constructions function
to identify an active zone when there is reason to make it explicit”. Some of
the constructions that can be used to indicate a temporally bounded relation
involving the profiled entity are When clauses (When he was a young man,
Joyce...), As + NP constructions (As a young man, Joyce...), and partitive
restrictive modification with an age adjective, among others. The important
Partitive restrictive modification of names in English...
45
aspect of these periphrases marking the active zone (Langacker, 1999: 65) is
that they make totally or partially explicit the covert, yet real “correspondence”
(that is the “mapping”) between a profiled entity and its active zone; in cases
like example (4), the active zone is made only partially explicit (for further
comment on the issue of explicitness, see 4.4).
The fact that zone activation is not easily noticeable does not mean that it is
not real. As Langacker claims, active zone / profile discrepancy is the normal
situation, and it is so common that it is hardly noticeable. The normal profile of
a name is a unitary and unique entity but this entity can be used as a metonymic
reference point to one of its possible active zones.
On two further objections to a metonymic analysis, see 4.4.
4.3. Partition into a figurative class and unique reference in reality. Analogy.
In partitive restrictive modification, the circumstantial relation constituting
the active zone of the name referent is always explicitly or implicitly contrasted
with others in the (sub)domain. In example (4), the active zone Joyce when
young implicitly contrasts with all other relationships involving Joyce within
the age domain (one of the set of domains constituting the conceptual network
or domain matrix for joyce’s biography, which is in turn one of the domain
matrices consituting the overall domain joyce), since age is the specific domain
(as a mental space) directly activated by the modifier (i.e. the spacebuilder)
young. An inherent component of age is a measurement scale with several
discrete points (the various ages in life), each of which in turn participates in
other conceptual networks (i.e. the set of experiences, social beliefs and other
types of knowledge socially shared in a given historical period and culture
about young, middle-aged and old people). The age domain is activated (and
through it the a mental space of Joyces’s biography) every time that, thanks to
a restrictive modifier (young, old, middle-aged) or thanks to other grammatical
means (e.g. In his youth / As a young man, Joyce...; In his old age, Joyce...), a
certain point on the age scale is mentioned:
(6) The young Joyce was already creative (active zone: joyce when young)
(7) The middle-aged Joyce wrote his first masterpiece (active zone: joyce when
middle-aged)
(8) The aging Joyce published his final masterpiece (active zone: joyce when
he was an aging man)
The circumstantial active zone also contrasts implicitly, but less
automatically, with other sub-domains in Joyce’s biography; one of these subdomains is affect and within it, love relationships:
46
Antonio Barcelona Sánchez
(9) The Joyce that married Nora Barnacle, an uneducated girl, was a dedicated
artist of the first order.
The active zone in (4) also contrasts implicitly, but even less automatically,
with other relationships outside his biography, in which Joyce might participate;
one of these might be the admiration for Joyce felt by the speaker:
(10) The Joyce that I admire most is the author of “Finnegans Wake”
The contrast between youth as an active zone and other circumstances,
biographical or otherwise, as possible active zones of Joyce, is only implicit
in (4) and in (9-10), and quite explicit in (11-13), the explicit contrast being
coded in each example by means of two instances of partitive restrictive
modification:
(11) The young Joyce had not reached yet the literary craftsmanship of the
mature Joyce.
(12) The young Joyce somehow contrasts with the Joyce that returned to Zurich
in 1940.
(13) The young Joyce is not the Joyce that I admire most.
See Figure 2 on this figurative class.
Figure 2: Figurative class of entities consisting of Joyce as involved in various
relationships. Based on metonymy in figure 1.
JOYCE
BIOGRAPHY
The young Joyce
The Joyce that wrote
Ulysses
The mature Joyce
The Joyce that returned
to Zurich in 1940
The Joyce that married
Nora Barnacle
CRITICAL
APPRAISAL
OF HIS
WORK
The
Joyce
that I
admire
most
Partitive restrictive modification of names in English...
47
All of these relationships (i.e. all of these possible active zones) constitute
a class, whose members are connected by the fact that they are all relationships
involving Joyce, the well known writer.
As has been repeatedly mentioned above, the active zones of Joyce in
example (4) and in examples (6-13) might have been cued grammatically by
other means. Why is partitive restrictive modification used, rather than the
other devices, and which is its effect?
In my view, there is one additional factor motivating the use of this
grammatical pattern: Its analogy with instances of non-partitive restrictive
modification on names, like the one in example (5) above (The young Joyce is
more intelligent than the older Joyce). Both in (4) and (5), two different entities
which are members of the same conceptual category by virtue of bearing the
same name are distinguished from each other by adding definite determiners
and restrictive modifiers to their common name. The only difference is that
in (5) the two different entities are two different unique individuals (i.e. two
different people), whereas in (4) the two different entities are two different
relations involving the same individual.
The effect of this grammatical pattern in (4) is to suggest the existence of a
figurative class of distinct individuals. This is why this modification is called
“partitive”. The partition is achieved by grammatically treating the different
highlighted relations in which the unitary entity participates (“Joyce as he
was young”, “Joyce as he was an aging man”, “Joyce as he wrote Ulysses”,
etc.), as a collection of different individuals. Yet, I would like to insist, the
ultimate conceptual motivation for the use of this construction is an active zone
metonymy (the applicability of this metonymic mapping depending in turn on
context and shared background knowledge).
The figurative partition achieved by means of partitive restrictive
modification seems to be an instance of what Fauconnier and Turner call
decompression, whereby a compressed (Fauconnier & Turner, 2002), compact
notion is figuratively split into some of its components or aspects.
What binds together this class of figuratively different individuals? The
fact that they are all, in the space of reality, only “aspects” (i.e. sub-domains)
of the same unitary individual (Joyce) and the fact that they are given the very
name (reclassified as a common noun) of that unitary individual. Once we have
a class we can make definite individuative reference to some of its members,
thanks to the narrowing down of the referential scope of the corresponding
NP that is jointly achieved by restrictive modifiers and definite determiners:
The young Joyce vs. The Joyce that returned to Zurich in 1940 (example 12).
A further example with definite individuative reference to a particular set of
entities:
48
Antonio Barcelona Sánchez
(14) All of these Joyces had one thing in common: they were extremely
obstinate.
(This sentence might come at the end of a paragraph discussing only James
Joyce the writer and mentioning, e.g. The young Joyce, The mature Joyce, and
The Joyce that married Nora)
In the realm of reality, there is only one unitary entity all the time in
reality; that is, in examples (4) and (6-14), the speaker really talks about one
and the same individual. This is one of the properties distinguishing names
with partitive restrictive modification from the use of names as paragons
(examples such Your friend is a real Shakespeare, listed in section 3 as use
c-2; see Barcelona, especially 2004). In paragons, the same name is given to
different unique individuals, which thus constitute a figurative class built on
the basis of a common ascribed property – “literary talent” in the Shakespeare
example). This is also the property that distinguishes partitive from nonpartitive restrictive modification (see example (5)): In the latter, two or more
unique individuals in reality which bear the same name are distinguished by
the properties picked out by the restrictive modifier (“young” vs “older” in
example 5).
The fact that the fictional individuals created by means of partitive
restrictive modification are, in reality space, just “aspects” or “circumstances”
of one and the same unitary individual inevitably constrains the functioning of
this figurative class as a real class of unique individuals. This can be shown by
the impossibility of expressing any type of generic reference to the fictional
class:
(15) *Joyces were born in Ireland
(16) *The Joyces were born in Ireland
(17) *The Joyce is a great writer (cf The Italian is normally extrovert)
(18) *A Joyce is obstinate (cf A student should pay attention to his teacher)
(referring in each case to the whole “Joyce class” constituted by “Joyce
when young”, “Joyce writing Ulysses”, “Joyce marrying Nora Barbacle”,
etc.)
4.4. On two objections on the role of metonymy
Two objections have been made to my claim regarding the role of metonymy
in partitive restrictive modification. These were made by two colleagues some
time ago in an exchange of messages on the topic. Here they are:
Partitive restrictive modification of names in English...
49
(a) One of them argued that, in an example like (4), the mental space activated
by the adjective young is that of someone’s youth, so that the expression The
young Joyce can readily yield the reading “Joyce when young” without any
metonymic account being necessary. Even though he did not explicitly say so,
this objection might be generalised as the claim that, if the discourse context
includes joyce’s biography as one the mental spaces currently active, then
young would be a spacebuilder activating one dimension (or “sub-space”) of
Joyce’s biography, namely, joyce’s youth, so that the fact that the NP The
young Joyce in the sentence The young Joyce already showed signs of the
genius that was to be fulfilled in Ulysses (= example 4) refers to “Joyce when
young” can be explained in a straightforward manner by appealing to the
activation of joyce’s youth within joyce’s biography. No metonymic account
would, then, be necessary.
(b) My other colleague argued that in the same example, the “meaning” of
the NP The young Joyce is roughly the same as that of Joyce when young (i.e.
that the two expressions are roughly synonymous). Therefore, the metonymic
target (Joyce when young) of the NP The young Joyce in that example is already
explicit. Since, according to his conception of metonymy, the metonymic
target must remain implicit, no metonymy could be proposed to account for
the meaning of the NP.
With respect to the first objection, I must reply that it is simply not true that
pointing out the activation of the mental space of youth by the adjective young
is enough to account for the reading of the NP The young Joyce in example
(4). That reading is “Joyce when young”, which adds a specific temporal
dimension to the meaning of the NP. The mere activation of youth does not
bring about the activation of the specific period when someone was young. It is
true that the fact that James Joyce’s youth extended over a specific span of time
is implicit in the mental space of joyce’s biography, which may be activated
through its youth “sub-space”; however, this fact is not highlighted or brought
to the reader’s or hearer’s attention only on the basis of the activation of that
mental space. Even in a discourse context in which joyce’s biography were
active, a sentence like (4) would not be necessarily interpreted as referring to
“Joyce when young”. It might also be interpreted as referring to “the younger
Joyce” (comparing him with his father or with an older relative with the same
surname); or even if the current discourse context is discussing Joyce’s work
during his youth, the sentence could still be interpreted as referring simply to
“Joyce who (by the way) was then young” (i.e. the modifier young would be
non-restrictive and non-partitive; cf. section 3, use d-2, example The fondly
remembered John F. Kennedy). In any of these contexts, we would still need
another factor leading to the temporally bound reading “Joyce when young”.
50
Antonio Barcelona Sánchez
This factor is the active zone metonymy mapping joyce onto joyce when
young. Note that I claimed above that the mapping is purely conceptual, i.e.
from the whole joyce frame to a temporally bound active zone relation (joyce
when young). This is why I claimed above that the metonymy is prelinguistic,
that is, it is purely conceptual though it motivates linguistic behavior (the
use of partitive restrictive modification on a name). On the other hand, the
objection is also levelled at my purported claim that the NP The young Joyce
was metonymically mapped onto joyce when young. But that is not exactly
my claim.
In sum:
First, even in a discourse in which the more specific mental space joyce’s
literary biography within joyce’s biography is active, the most likely
interpretation of the NP The young Joyce would be joyce when young, but
the other interpretations would not be completely discarded. We would still
need the active zone mapping. Second, in discourse contexts which have not
activated joyce’s biography prior to the occurrence of this sentence, the active
zone metonymy is even more necessary both to produce and to interpret this
sentence (for example, in a debate on literary creativity in general in which
there had been no mention or allusion to Joyce yet, one of the participants
might utter sentence (4) to suggest indirectly that Joyce was always highly
creative –from his early work).
With respect to the second objection, first of all my claim is not that The young
Joyce maps onto joyce when young; I claimed that the frame joyce is mapped
onto the conceptual relation joyce when young, as explained a few lines above.
Secondly, the “meaning” of the NP The young Joyce is not roughly the same
as that of Joyce when young, given the multiple ambiguity of The young Joyce
(see my reply to the first objection) and given the temporal dimension of Joyce
when young. Thirdly, in any case, I never claimed that a linguistic expression
is metonymically mapped onto another linguistic expression; I claimed that a
concept (in this case, a frame) is mapped onto another concept (in this case, a
temporally bounded relation). Fourthly, the target of the metonymy is not fully
explicit; it is at most only partially explicit: The source is Joyce and the target
is the conceptual relation Joyce when young, as stated a few lines earlier. Even
if one regarded the metonymy that I advocate as a linguistic metonymy (i.e.
as a connection between a linguistic expression and its associated meaning, on
the one hand, and another linguistic expression and its associated meaning, on
the other hand), the temporal dimension contributed by when is absent from
the source. But the metonymy I propose is, not a metonymy of that sort. As just
stated above, it is a purely conceptual metonymy and it links two conceptual
entities, not two linguistic expressions. This conceptual metonymy, though,
Partitive restrictive modification of names in English...
51
motivates in part a certain linguistic behavior, consisting in this case in the use
of the definite article and the partitive restrictive modifier young attached to a
proper name like Joyce (as stated earlier in 4.2, the partial explicitness of the
active zone is necessary because it provides information otherwise unavailable).
It does seem that the target of prototypical linguistic metonymies is normally
implicit, so that implicitness might be established as requirement for these
types of metonymy..But the requirement of implicitness for purely conceptual,
pre-linguistic metonymies simply makes no sense, because the mapping
in these cases is not direcly “visible” (it is prelinguistic). This requirement
makes sense for linguistic metonymies, where a grammatical construction
(typically a lexical item) undergoes a directly observable meaning shift due
to metonymy, even if the target sense remains implicit (a shift that may even
become conventionalized); cf. shift in the meaning of hand in We need five
extra hands at the factory (“hand”-“manual worker”). But the metonymic
understanding of mother as a “housewife mother” (Lakoff, 1987) remains at
the level of conceptualization and is only revealed discursively in reasoning
(She’s a good mother though she has a demanding job as an executive), not
in an extended (implicit) sense of the lexeme mother. And the use of partitive
restrictive modification of names depends on the prelinguistic mapping of the
whole conceptual network (the frame) associated to the standard referent of the
name, Joyce in this case, onto one of its active zones: The metonymy joyce →
joyce when young determines the choice of the grammatical pattern The young
Joyce in example (4); see also section 4.2.
5. Summary and conclusions
This paper is intended as a contribution to the study of the role of metonymy
in grammar. After briefly surveying recent research on the role of metonymy
in the motivation of grammar, I have listed the main types of special uses of
names due to their transient reclassification as common nouns. One of these
types is the use of partitive restrictive modification on names, the topic of the
paper. It has been shown that this construction reflects a figurative splitting of
the referent into a class of referents consisting of a number of relations in which
the referent may be involved, even though the referent in reality continues to
be unique (we refer to the same individual by means of the two mentions of
Joyce in examples 11-13 above).
It has been argued that this construction is motivated by: (a) A purely
conceptual, prelinguistic metonymy mapping the whole cognitive frame
associated with the unique referent of the name onto one of its active zones
consisting in a relation involving that referent; this metonymy motivates the
52
Antonio Barcelona Sánchez
aforesaid figurative splitting of the referent into a class and the use of the
construction to code the splitting. And (b) the analogy of this construction
to instances of non-partitive restrictive modification on names; this type of
modification is a means (together with the definite determiners) of achieving
definite reference to the various members of the figurative class.
I have finally answered two objections claiming that metonymy plays no
role in the motivation of this construction. Among other things, I have shown
that the reading “Joyce when (he was) young” in example (4) is motivated
in part by an active zone metonymy because the past temporal dimension
(“when”) is not activated simply through the activation of the youth mental
space or even through the activation of the whole of joyce’s biography mental
space: The metonymic account is still necessary. And I have argued that in the
metonymy proposed (joyce for joyce when young) the target is not fully but
only partially explicit (in any case the requirement of implicitness does not
apply to purely conceptual, prelinguistic metonymies).
One of the most interesting aspects of this research is that it provides further
evidence (collected through the research on names by the author and other
linguists) for the claim that names are not merely unique designators of entities
whose grammatical behavior remains unaffected by the conceptual networks
associated with their referents. Quite the opposite, the grammatical behavior of
names is constantly governed by our rich knowledge network of their referents
(see Michaux, 1998).
6. References
Barcelona Sánchez, A. (2002). “Clarifying and applying the notions of
metaphor and metonymy within cognitive linguistics: An update”. In:
Dirven & Pörings, 207-277.
Barcelona Sánchez, A. (2003a). “Metonymy in cognitive linguistics: An
analysis and a few modest proposals”. In: H. Cuyckens, T. Berg, R. Dirven
& K.-U. Panther (eds.) (2003), 223-255.
Barcelona Sánchez, A. (2003b). “La gramática de los nombres propios: un
viaje con retorno (por buenos motivos)”. In: M. White, H. Herrera & C.
Alonso (eds.), Cognitive Linguistics in Spain at the turn of the Century
(II). (Metaphor and Metonymy)/ La Lingüística Cognitiva en España en
el cambio de Siglo (II). (Metáfora y Metonimia). Madrid: Universidad
Autónoma de Madrid, 7-30.
Partitive restrictive modification of names in English...
53
Barcelona Sánchez, A. (2003c). “Names: A metonymic return ticket”.
Jezikoslovlje 4.1. Osijek, Croatia: Josip Juraj Strossmayer University, 1141.
Barcelona Sánchez, A. (2004). “Metonymy behind grammar: The motivation
of the seemingly ‘irregular grammatical behavior of English paragon
names”. In: Radden & Panther, 357-374.
Barcelona Sánchez, A. (2008). “The interaction of metonymy and metaphor
in the meaning and form of ‘bahuvrihi’ compounds”. Annual Review of
Cognitive Linguistics 6 (6): 208-281.
Barcelona Sánchez, A. (2009). “Metonymy in constructional meaning and form:
Its motivational and inferential roles”. In: K.-U. Panther, L. Thornburg &
A. Barcelona Sánchez (eds.), 363-401.
Barcelona Sánchez, A. (In preparation a). On the Pervasive Role of Metonymy
in Constructional Meaning and Structure in Discourse Comprehension: An
Empirical Study from a Cognitive-Linguistic Perspective (provisional title).
[Cognitive Linguistic Research]. Berlin/New York: Mouton de Gruyter.
Barcelona Sánchez, A. (In preparation b). “Metonymy-motivated
morphosyntactic ‘alternations’”. In: P. Guerrero, Morphosyntactic
alternations (provisional title). Córdoba: University of Córdoba.
Barcelona Sánchez, A. (n. d.). “Metonymy behind grammar: The motivation
of the grammar of English paragon and modified names”. Unpublished ms.
University of Murcia, Spain
Basilio, M. (2009). “The role of metonymy in word formation: Brazilian
Portuguese agent noun constructions”. In: K.-U. Panther, L. Thornburg &
A. Barcelona Sánchez (eds.).
Benczes, R., A. Barcelona Sánchez & F. J. Ruiz de Mendoza Ibáñez (eds.)
(2009). What is metonymy? An attempt at building a consensus view on the
delimitation of the notion of metonymy in Cognitive Linguistics (provisional
title). (Submitted for publication to John Benjamins, Human Cognitive
Processing Series).
Brdar, M. (2007a). “When Zidane is not simply Zidane, and Bill Gates is
not just Bill Gates: Some thoughts on the construction of metaphtonymic
meanings”. In: G. Radden, K.-M. Koepcke, T. Berg & P. Siemund
(eds.), Aspects of Meaning Construction. Amsterdam/Philadelphia: John
Benjamins, 125-142.
Brdar, M. (2007b). Metonymy in Grammar. Towards Motivating Extensions of
Grammatical Categories and Constructions. Osijek (Croatia): Faculty of
Philosophy, Josip Juraj Strossmayer University.
54
Antonio Barcelona Sánchez
Brdar, M. & R. Brdar-Szabó (2004). “Predicative adjectives and grammaticalrelational polysemy: The role of metonymic processes in motivating crosslinguistic differences”. In: Radden & Panther, 321-355.
Brdar, M. & R. Brdar-Szabó (2009). “The (non-)metonymic use of place
names in English, German, Hungarian, and Croatian”. In: K.-U. Panther,
L. Thornburg & A. Barcelona Sánchez (eds.), 229-257.
Brdar-Szabó, R. (2009). “Metonymy in indirect directives: Stand-alone
conditionals in English, German, Hungarian, and Croatian”. In: K.-U.
Panther, L. Thornburg & A. Barcelona Sánchez (eds.), 323-336.
Caballero, R. (2009). “FORM IS MOTION: Dynamic predicates in English
architectural discourse”. In: K.-U. Panther, L. Thornburg & A. Barcelona
Sánchez (eds.), 277-290.
Clark, E. V. & H. H. Clark (1979). “When nouns surface as verbs”. Language
55: 767-811.
Dirven, R (1999). “Conversion as a conceptual metonymy of event schemata”.
In: K.-U. Panther & G. Radden, Metonymy in cognition and language.
Amsterdam/Philadelphia: John Benjamins, 275-287.
Dirven, R. & R. Pörings (eds.) (2002). Metaphor and Metonymy in Comparison
and Contrast [Cognitive Linguistics Research 20]. Berlin: Mouton de
Gruyter.
Fauconnier, G. (1997). Mappings in Thought and Language. Cambridge:
Cambridge University Press.
Fauconnier, G. & M. Turner (2002). The Way we Think. New York: Basic
Books.
Fillmore, Ch. (1985). “Frames and the semantics of understanding”. Quaderni
di Semantica 6: 222-254.
Greenbaum, S. & R. Quirk (1990). The student’s grammar of the English
language. London: Longman.
Guarddon, C. (In preparation). “Metonymy and metaphor in grammaticalization:
Two aspectual prefixes in Polish”. In: K.-U. Panther, L. Thornburg & A.
Barcelona Sánchez (eds.).
Jespersen, O. (1949). A modern English grammar on historical principles. Part
VII: Syntax. (Completed and edited by Niels Haislund.) London: George
Allen and Unwin Ltd and Copenhaguen: Ejnar Munksgaard.
Kövecses, Z. & G. Radden (1998). “Metonymy: Developing a cognitive
linguistic view”. Cognitive Linguistics 9.1: 37-77.
Lakoff, G. (1987). Women, fire and dangerous things. What categories reveal
about the mind. Chicago: Chicago University Press.
Lakoff, G. & M. Turner (1989). More than cool reason: A field guide to poetic
metaphor. Chicago: University of Chicago Press.
Partitive restrictive modification of names in English...
55
Langacker, R. W. (1987). Foundations of Cognitive Grammar. Vol. I:
Theoretical Prerequisites. Stanford: Stanford University Press.
Langacker, R. W. (1991). Concept, Image and Symbol: The Cognitive Basis of
Grammar [Cognitive Linguistics Research 1]. Berlin/New York: Mouton
de Gruyter.
Langacker, R. W. (1993). “Reference-point constructions”. Cognitive
Linguistics 4, 1-38.
Langacker, R. W. (1995). “Raising and transparency”. Language 71: 1-62.
(Adapted, with substantial revisions, as chapter 11 of Langacker, 1999).
Langacker, R. W. (1999). Grammar and conceptualization. Berlin/New York:
Mouton de Gruyter.
Langacker, R. W. (2009). “Metonymic grammar”. In: K.-U. Panther, L.
Thornburg & A. Barcelona Sánchez (eds.), 45-71.
Michaux, Ch. (1998). “Dénominations et répresentations conceptuelles”. In: J.
L. Cifuentes (ed.). Estudios de lingüística cognitiva II. Alicante: University
of Alicante, Spain, 755-765.
Palmer, G.; R. S. Rader & A. D. Clarito (2009). “The metonymic basis of
a ‘semantic partial’: Tagalog lexical constructions with ka-”. In: K.-U.
Panther, L. Thornburg & A. Barcelona Sánchez (eds.), 111-144.
Panther, K.-U. & L. Thornburg (2000). “The Effect for Cause Metonymy in
English Grammar”. In: A. Barcelona Sánchez (ed.), Metaphor and metonymy
at the crossroads. A cognitive perspective. Berlin/New York: Mouton de
Gruyter, 214-231.
Panther, K.-U. & L. Thornburg (2002). “The roles of metaphor and metonymy
in English –er nominals”. In: Dirven & Pörings, 279-319.
Panther, K.-U. & L. Thornburg (2009). “Aspect and metonymy in the French
passé simple”. In: K.-U. Panther, L. Thornburg & A. Barcelona Sánchez
(eds.), 117-195.
Panther, K.-U., L. Thornburg & A. Barcelona Sánchez (eds.) (2009). Metonymy
and Metaphor in Grammar. (Human Cognitive Processing.). Amsterdam/
Philadelphia: John Benjamins.
Paradis, C. (2004). “Where does metonymy stop? Senses, facets and active
zones.” Metaphor and Symbol, 19(4): 245-264.
Paradis, C. (In preparation). “Metonymization: key mechanism in semantic
change”. In: R. Benczes, A. Barcelona Sánchez & F.-J. Ruiz de Mendoza
Ibáñez (eds.).
Peirsman, Y. & D. Geeraerts (In preparation). “Zones, facets and prototypebased metonymy”. In: R. Benczes, A. Barcelona Sánchez & F.-J. Ruiz de
Mendoza Ibáñez (eds.).
56
Antonio Barcelona Sánchez
Peña Cervel, M. S. & F. J. Ruiz de Mendoza Ibáñez (2009). “The metonymic
and metaphoric grounding of two image-schema transformations”. In:
K.-U. Panther, L. Thornburg & A. Barcelona Sánchez (eds.), 339-361.
Quirk, R., S. Greenbaum, E. Leech & J. Svartvik (1985). A Comprehensive
Grammar of the English Language. London: Longman.
Radden, G. (2005). “The ubiquity of metonymy”. In: J. L. Otal Campo, I.
Navarro i Ferrando & F. Begoña Bellés (eds.). Cognitive and Discourse
Approaches to Metaphor and Metonymy, 11-28. Castellón: Jaume I
University.
Radden, G. (2009). “Generic reference: An instance of the instance for type
metonymy”. In: K.-U. Panther, L. Thornburg & A. Barcelona Sánchez
(eds.), 199-228.
Radden, G. & R. Dirven (2007). Cognitive English Grammar. (Cognitive
Linguistics in Practice.) Amsterdam & Philadelphia: John Benjamins.
Ruiz de Mendoza, F. J. & J. L. Otal Campo (2002). Metonymy, Grammar and
Communication. Albolote (Granada, Spain): Comares.
Traugott, E. C. & R. B. Dasher (2002). Regularity in Semantic Change.
Cambridge: Cambridge University Press.
Ziegeler, D., & S. Lee (2009). “A metonymic analysis of Singaporean and
Malaysian English causative constructions”. In: K.-U. Panther, L.
Thornburg & A. Barcelona Sánchez (eds.), 291-322.